Staff Reports
Bank Leverage Limits and Regulatory Arbitrage: New Evidence on a Recurring Question
Previous title: “Leverage Limits and Bank Risk: New Evidence on an Old Question”
Number 856
June 2018 Revised November 2018

JEL classification: G20, G21, G28

Authors: Dong Beom Choi, Michael R. Holcomb, and Donald Morgan

Bankers are regulated more than most business owners, giving them more incentive to arbitrage (avoid) regulation via loopholes. The supplementary leverage ratio (SLR) rule may present their latest arbitrage opportunity. Enacted after the crisis to prevent another leverage buildup, the rule caps leverage at the very largest U.S. banks. While the leverage rule is simpler than risk-based capital requirements because it requires equal capital against assets with unequal risk, bankers can arbitrage by shedding safer assets and/or adding riskier ones. An earlier leverage rule imposed in 1981 invited the same arbitrage, but there is little or no evidence that bankers exploited it. We find more compelling evidence of leverage rule arbitrage around the new rule. Studying difference-in-differences, we find higher risk (risk-weighted) asset shares and security yields at SLR banks relative to the control (the next largest set of banks) after the SLR was finalized in 2014. The effects tend to be larger at more leverage-rule-constrained banks, and some are substantial; mean yields at SLR banks rose (relatively) about 30 basis points. While this arbitrage might have, perversely, increased overall bank risk, we find no evidence that it did. Book and market risk measures were all essentially unchanged except one: leverage. The most constrained SLR banks significantly de-levered, commencing precisely when public disclosure of leverage ratios was required. Their reduced leverage may have offset the asset arbitrage, leaving overall risk unchanged.

Available only in PDF
AUTHOR DISCLOSURE STATEMENT(S)
Dong Beom Choi
I am employed by the Federal Reserve Bank of New York. The views expressed in the article under submission reflect my views and the views of my co-authors, but do not necessarily represent the views of the Federal Reserve Bank of New York, Federal Reserve Board, or the Federal Reserve System.
I have not received outside financial support for the research in this article.
I have not received any fees or payments from any institutions that might be relevant to the content of the research under submission.
No close relative has received funding or financial support, or is an officer, director, or board member of any relevant organization.
The article was vetted prior to circulation outside the Federal Reserve Board and the Federal Reserve Bank of New York.

Michael R. Holcomb
The author declares that he has no relevant or material financial interests that relate to the research described in this paper.

Donald P. Morgan
I, Donald P. Morgan, have no relevant or material financial interests that relate to the research described in this paper.